Re: [RFC PATCH -rt 2/2] RCU priority boosting additions to rcutorture

["Paul E. McKenney" <paulmck@linux.vnet.ibm.com>]

On Thu, Feb 01, 2007 at 01:12:16PM +1100, Nigel Cunningham wrote:
> Hi Paul.
> 
> On Wed, 2007-01-31 at 17:26 -0800, Paul E. McKenney wrote:
> > This patch adds an optional preemption kernel thread to the rcutorture
> > tests.  This thread sets itself to a low RT priority and chews up CPU
> > in 10-second bursts, verifying that grace periods progress during this
> > 10-second interval.  Passes RCU torture testing on a 4-CPU (a pair of
> > 2-CPU dies) 64-bit Xeon system.
> 
> [...]
> 
> > Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
> > +static int rcu_torture_preempt(void *arg)
> > +{
> > +	int completedstart;
> > +	int err;
> > +	time_t gcstart;
> > +	struct sched_param sp;
> > +
> > +	sp.sched_priority = MAX_RT_PRIO - 1;
> > +	err = sched_setscheduler(current, SCHED_RR, &sp);
> > +	if (err != 0)
> > +		printk(KERN_ALERT "rcu_torture_preempt() priority err: %d\n",
> > +		       err);
> > +	current->flags |= PF_NOFREEZE;
> > +
> > +	do {
> > +		completedstart = rcu_torture_completed();
> > +		gcstart = xtime.tv_sec;
> > +		while ((xtime.tv_sec - gcstart < 10) &&
> > +		       (rcu_torture_completed() == completedstart))
> > +			cond_resched();
> > +		if (rcu_torture_completed() == completedstart)
> > +			rcu_torture_preempt_errors++;
> > +		schedule_timeout_interruptible(HZ);
> > +	} while (!kthread_should_stop());
> > +	return 0;
> > +}
> 
> Does it need to be NOFREEZE? I would think that it should be frozen
> during a suspend/hibernate.

Good to hear from you, Nigel!

Should indeed be OK to freeze during suspend/hibernate.  Will my
schedule_timeout_interruptible() be sufficient to allow the freeze
to happen, or do I need to add an explicit try_to_freeze()?

Ah, and I probably need to use the same trick that mtd_blktrans_thread()
does to avoid having all my sleeps killed of by an errant signal:

	spin_lock_irq(&current->sighand->siglock);
	sigfillset(&current->blocked);
	recalc_sigpending();
	spin_unlock_irq(&current->sighand->siglock);

Or is such paranoia unnecessary?

							Thanx, Paul
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/